Many of us educators, in our roles as parents, have worked hard to get our own children into the classroom of an unusually good teacher—or out of the classroom of a teacher who does serious damage. But interestingly, in our roles as educators, we often deny that such differences exist. When those pesky parents ask us to assign their children to a particular teacher's classroom, what do we say? "Oh, don't worry. Your child will learn what she needs to learn from any teacher in our school."

Well, that turns out to be a lie. There are big differences in the amounts and kinds of learning that different teachers help produce. As a study (Gordon, Kane, & Staiger, 2006) in Los Angeles showed, students taught by teachers in the top quartile of effectiveness advance, on average, approximately five percentile points each year relative to their peers, whereas those taught by teachers in the bottom quartile of effectiveness lose, on average, five percentile points relative to their peers. Moreover, these effects are cumulative. The same study suggested that if all black students were assigned to four highly effective teachers in a row, this would be sufficient to close the average black-white achievement gap.

Who's Teaching Whom?

So teachers are hugely important. But no matter how you measure quality, good teachers are not evenly distributed across all kinds of schools and students.

For years we've had only proxies for teacher quality, and often not strong ones at that. We've known whether teachers are licensed or not, are teaching in-field or not, and are experienced or novices. Sometimes we've known how they performed in college or on licensure tests.

Year after year, decade after decade, countless studies told us that on these measures, we didn't have a fair distribution of teacher talent (Darling-Hammond, 1995; Kain & Singleton, 1996; Lankford, Loeb, & Wyckoff, 2002; Presley, White, & Gong, 2005; Shields et al., 1999). Minority and poor students in particular were typically taught by significantly more than their fair share of unlicensed, out-of-field, and inexperienced teachers who often didn't have records of strong academic performance themselves.

However, these data were often dismissed as being of questionable import. Are unlicensed teachers really worse than others? Aren't most private school teachers uncertified, and aren't Teach for America corps members unlicensed—and are they so bad? As for this matter of experience, we all know teachers who are better in year two of their practice than others are in year 20. So who are we to moan about statistics suggesting that teachers in high-poverty schools are less experienced than those working in more-affluent schools?

People in and around schools might feel this way about the data, but the U.S. Congress didn't share that ambivalence. "Wait a second," said congressional leader George Miller (D-CA). "We're giving these schools funds to provide extras for low-income students, but they keep shorting these kids in the thing they most need—quality teachers." So in 2001, Congress asked states to do something about the problem: to revisit their requirements for teachers, decide what would prove that a teacher was "highly qualified," and then eliminate gaps in teacher qualification.

Unfortunately, most states decided to label everybody as highly qualified, and the U.S. Department of Education turned a blind eye to this practice. So the United States is pretty much in the same position today that it's been in for decades. We have data that strongly suggest that teachers in high- and low-poverty schools differ greatly on a wide range of professional characteristics, but we have incomplete information on how much those characteristics matter—and a weak commitment to doing anything about it.

What the Data Show

Fortunately, some states and school districts are in a much better position to understand the nature and extent of differences in teacher quality. Tennessee is perhaps the best positioned of all because, in addition to collecting the usual data about certification status, experience, and licensure performance, the state has for more than a decade produced "value-added" data for most of its teachers. In brief, this approach averages multiple years of data for every teacher and compares the growth the students of those teachers make with the growth made by other students in the same grade and subject. These data are adjusted according to the previous trajectory of every student, providing an even playing field for evaluating the effect of the state's teachers on student learning. Some teachers consistently produce much larger gains than the average, whereas others produce much smaller gains.

Recently, the state education department in Tennessee took a close look at what these data told them about the kinds of students being taught by the strongest—and weakest—teachers in the state. Unfortunately, what they found echoes the patterns in the proxy data: Poor children and black children were less likely to be taught by the strongest teachers and more likely to be taught by the weakest.

This shouldn't have been a surprise. Earlier data from Tennessee had suggested a similar pattern. Also, the Dallas Independent School District—which has computed similar value-added data—has consistently reported serious disproportions in the kinds of students taught by high- and low-value-added teachers (Babu & Mendro, 2003).

The bottom line is clear. Yes, poor and minority students often enter school behind other students. But rather than organizing our system to ameliorate that problem, we exacerbate it by assigning them disproportionately to our least effective teachers.

What Schools Are Doing

In our own work on thorny issues, we typically start by trying to learn from the schools or districts that are already tackling a given problem with some success. So let's look at several places—some that are just beginning and some well established—that have implemented programs to reduce the teacher quality gap. These initiatives and programs exemplify what we can do when we are serious about providing strong teachers to the students who most need them.

Chattanooga's Benwood Initiative

Hamilton County, Tennessee, has a well-established program. The county school district's Benwood Initiative was launched after a statewide report identified nine of Hamilton County's elementary schools—all in Chattanooga—as among the 20 lowest-achieving schools in the state. The initiative, a combined effort of the Benwood Foundation, the local Public Education Fund, and the school district, focuses on improving achievement in these nine schools by improving the quality of instruction.

Their first step was to recognize that without strong leadership, the schools would not meet the aggressive goal of getting all 3rd graders to grade level in reading by 2006 (Chenoweth, 2007). So the district hired new principals for the majority of the schools, and the Public Education Fund started a principal leadership institute in which principals learned how to use data and coach teachers to improve instruction.

For these principals to truly focus on instruction, however, Superintendent Jesse Register knew that he had to remove concentrations of ineffective teachers. Over the years, Benwood schools had become the dumping ground for teachers whom other district schools did not want. The district decided to reconstitute all nine Benwood schools, asking all teachers to reapply for their jobs. Superintendent Register asked principals at surrounding suburban schools to take one weak teacher from the 28 tenured, low-performing teachers let go from Benwood schools who could not find jobs at other schools (Handley & Kronley, 2006).

With new leadership and staff in place, the district implemented job-embedded professional development to improve teachers' instructional practice. Key to the new approach were consulting teachers hired to work with the staff at the Benwood schools. In addition, specially selected teachers became Osborne Fellows, a status that entitled them to full scholarships in master's degree programs customized to fit the needs of the district and the unique challenges of urban teaching. The Osborne Fellows became leaders in their schools to help improve the quality of instruction.

The final piece of the district's strategy was to use data from the Tennessee Value-Added Assessment System to assess the growth of student learning in the classrooms of the teachers at the nine Benwood Schools, rewarding teachers with bonuses of $5,000 when their students' scores on the state achievement test were at least 15 percent higher than the amount of expected growth. The results of the Benwood initiative? Benwood schools went from 53 percent of their 3rd graders scoring at the advanced or proficient level in reading on the Tennessee Comprehensive Assessment Program to 80 percent scoring at that level in 2007.

New York's Teaching Fellows Program

For many years, the New York City public schools suffered from serious teacher-quality problems in the city's highest-poverty schools. Many teachers weren't even licensed; others had failed the licensure exam on multiple occasions.

Finally, Rick Mills, the state education commissioner, ordered the city to stop hiring unlicensed teachers. Instead of doing the bare minimum, however, district leaders invited the New Teacher Project, a nonprofit organization with a track record of recruiting strong candidates for classroom teaching, to help them design and launch a new NYC Teaching Fellows program to attract accomplished professionals from education and other fields to become teachers. This involved advertising aggressively, emphasizing both the challenge of teaching in high-poverty settings and the enormous opportunity to make a difference. By 2007, teaching fellows represented 10 percent of teachers in New York City (New Teacher Project, 2007).

In 2004, the New Teacher Project partnered with the district to help improve hiring and placement practices more generally. The organization provided crucial data that pointed to a serious problem with "must-hire" teachers—teachers who had been removed from one school (often for performance reasons) and who were then forced on another school (often one with high poverty). The district adopted new transfer rules, which eliminated transfers solely on the basis of seniority and allowed principals to select teachers on the basis of their match with the school (New Teacher Project, 2007).

A recent external study has shown significant positive effects from this collaboration, especially on the characteristics of new teachers. In 2000, new teachers in high-poverty schools in New York City were far more likely to have failed their licensure exam than their counterparts in low-poverty schools. By 2005, however, new teachers in high-poverty schools were actually less likely to have failed the exam (Boyd, Grossman, Lankford, Loeb, & Wyckoff, 2007).

Boston, Chicago, and Colorado: Teacher Residency Programs

Frustrated by the small number of talented new teachers from traditional university-based teacher preparation programs who want to teach in high-poverty schools, Boston, Chicago, and two systems in Colorado have taken matters into their own hands, establishing their own teacher residency programs modeled on medical residencies.

Using tough selection criteria, teacher residency programs hire recent college graduates, career changers, and others willing to make a multiyear commitment to teaching in some of the most challenging schools in the districts. The Boston Teacher Residency requires a four-year commitment; the Academy for Urban School Leadership in Chicago requires a commitment of six years. Recruitment strategies vary for each residency. For example, the Boston Teacher Residency reaches out to local entities such as churches and civic organizations and places advertisements in community publications and on public transportation. The program also hires college students to recruit on college campuses.

In both programs, residents do not spend their first year as the teacher of record. Instead, they are assigned to work under a mentor teacher, spending part of their week in that teacher's classroom and at least one day each week taking courses. Residents are paid a stipend rather than a full teacher salary. At the end of their 12- to 13-month residencies, residents in the Boston Teacher Residency and the Academy for Urban School Leadership typically have earned enough credits for a master's degree; participants in the Boettcher Teachers Program in Colorado finish their graduate work in their second year. As residents move into full-time teaching positions, they continue to receive support from a mentor.

The Academy for Urban School Leadership started in 2001 and has the longest-running residency program. It boasts a retention rate of 95 percent and has 153 graduates serving more than 4,500 low-income students in Chicago (Academy for Urban School Leadership, 2008). The Boston Teacher Residency, which launched in 2003, is currently training 84 teachers. So far, the program has 140 graduates teaching in Boston Public Schools; it has a retention rate of 90 percent, compared with the district's new-teacher retention rate of 53 percent. The Boettcher Teachers Program is having similar success. The program has placed 55 teachers in target districts serving low-income and minority students.

Steps to Take Right Now

Programs like these show us a general approach to providing more high-quality teachers to students who most need them.

• Start with your data. Although you may not have access to perfect data on the characteristics of teachers in different kinds of schools, most school districts keep data that enable them to look at things like turnover, out-of-field teaching, and years of teaching experience. Compare the statistics for high- and low-poverty schools and high- and low-minority schools.
• Ask the question, If we don't change these patterns, can we honestly claim we are doing everything we can to close long-standing achievement gaps? Have an honest conversation about this with your teacher and administrator colleagues and with your local school board. Gaps in teacher quality are a big contributor to gaps in student achievement.
• Start a conversation on how to attract strong teachers to the schools that are not getting their fair share. Ask some of the strongest teachers in the district what it would take to get them to teach in one of the most challenging schools. Some will talk about incentive pay, some about reduced class size, some about better principals, and some about "better" colleagues. But sometimes the answers are more surprising. "They could start by fixing the bathrooms," one teacher told us not long ago.
• Don't neglect within-school differences. In addition to between-school differences in teacher quality, there are often even bigger within-school differences. At the high school level, for example, 9th graders—actually the most vulnerable of students—are more likely to be taught by out-of-field and inexperienced teachers than are 12th graders. At every level, low-achieving students—the very students who most need strong teachers—are the least likely to be assigned them. Some principals find that when they ask their most accomplished teachers to take on this challenge, the teachers readily agree. Others find that it takes a carrot—such as an extra prep period—or a change in policy—such as requiring teachers who teach advanced courses to also teach low-performing 9th graders.
• Design and carry out a multipronged strategy. No single strategy will work. Most systems are trying some combination of better principals, salary incentives for strong teachers to stay in or move into high-poverty schools, subsidized master's programs or other professional development, and significantly reduced class sizes. An increasing number of districts are "growing their own" teachers for high-poverty schools.
• Improve your data system to help you make better decisions and evaluate their effects. Most systems have plenty of data to start this work. Over time, those schools that have not yet linked teacher and student data to produce "value-added" estimates of school and teacher effects will want to do this. It's helpful to teachers as well as to school and district leaders who want to analyze what does and doesn't work in improving teacher effectiveness.

Putting Effort Where It Counts

None of these steps is easy. Even if you act aggressively, you will not reverse long-standing patterns of teacher distribution overnight.

But most of us got into this profession to make a difference. We've known for years that education has real power to change lives for the better. We know now that teachers have enormous transformative power, too. It's time we step up and harness that power to help close the gaps that consign so many of our students to lives on the margins.

References

Academy for Urban School Leadership. (2008). About us. Chicago: Author. Available: www.ausl-chicago.org/about.html

Babu, S., & Mendro, R. (2003, April). Teacher accountability: HLM-based teacher effectiveness indices in the investigation of teacher effects on student achievement in a state assessment program. Paper presented at the American Educational Research Association meeting, Chicago.

Boyd, D., Grossman, P., Lankford, H., Loeb, S., & Wyckoff, J. (2007). Who leaves? Teacher attrition and student achievement. Albany, NY: Teacher Pathways Project. Available: www.teacherpolicyresearch.org/portals/1/pdfs/Do%20Effective%20Teachers%20Leave.pdf

Chenoweth, K. (2007). "It's being done": Academic success in unexpected schools. Cambridge, MA: Harvard Education Press.

Darling-Hammond, L. (1995). The role of teacher expertise and experience in students' opportunity to learn. In P. Brown (Ed.), Strategies for linking school finance and students' opportunity to learn (pp. 19–23). Washington, DC: National Governors Association.

Gordon, R., Kane, T. J., & Staiger, D. O. (2006). Identifying effective teachers using performance on the job. Washington, DC: Brookings Institution.

Handley, C., & Kronley, R. A. (2006). Challenging myths: The Benwood Initiative and education reform in Hamilton County. Chattanooga, TN: Public Education Foundation. Available: www.pefchattanooga.org/Portals/0/Benwood/challengingmyths.pdf

Kain, J. F., & Singleton, K. (1996, May/June). Equality of educational opportunity revisited. New England Economic Review, 87–114.

Lankford, H., Loeb, S., & Wyckoff, J. (2002). Teaching sorting and the plight of urban schools: A descriptive analysis. Educational Evaluation and Policy Analysis, 24(1), 37–62.

New Teacher Project. (2007). Impact: New York City. New York: Author. Available: www.tntp.org/ourimpact/impact_nyc.html

Presley, J., White, B., & Gong, Y. (2005). Examining the distribution and impact of teacher quality in Illinois. Edwardsville: Illinois Education Research Council. Available: http://ierc.siue.edu/documents/Teacher%20Quality%20IERC%202005-2.pdf

Shields, P. M., Esch, C. E., Humphrey, D. C., Young, V. M., Gaston, M., & Hunt, H. (1999). The status of the teaching profession: Research findings and policy recommendations. Santa Cruz, CA: Center for the Future of Teaching and Learning.

---

Kati Haycock (khaycock@edtrust.org) is President and Candace Crawford (ccrawford@edtrust.org) is Senior Associate for Teacher Quality, The Education Trust, Washington, D.C. Haycock also serves as Chair of the Board for the New Teacher Project.